Viewing OS Metrics

Cloud Eye collects OS metrics from your ECSs or BMSs through their installed Agents. You can configure alarm rules based on these metrics. When a metric reaches the specified alarm threshold, Cloud Eye will send you an alarm notification. Cloud Product Metrics lists the OS monitoring metrics supported by Cloud Eye.

Differences Between Basic Monitoring and OS Monitoring

Dimension	Basic Monitoring	OS Monitoring
Monitoring location	Cloud server	Cloud server and OS
Collection frequency	Most metrics are collected every 5 minutes and reported by cloud services.	1 minute. Data is proactively collected.
Installation requirements	No Cloud Eye Agent needs to be installed.	The Cloud Eye Agent is required.
Advantage	Wide application scope	High data precision
Disadvantage	Low precision, unable to identify sudden metric data fluctuations.	The maintenance Agent is required.
Scenario	General monitoring scenarios that monitor the overall usage of server hardware and resources	Advanced O&M scenarios that require in-depth understanding of system operations

Prerequisites

You have installed the Agent. For details, see Installing the Agent.

Viewing OS Metrics

Perform the following operations based on your resources.
- In the navigation pane, choose Server Monitoring > Elastic Cloud Server.
- In the navigation pane, choose Server Monitoring > Bare Metal Server.
Locate the target server and click View Metric in the Operation column.
On the OS Monitoring page, view the graphs of the current monitored instance.
Figure 1 OS monitoring
In the upper part, choose Compute, Storage, Networking, Time Synchronization, Accelerator Cards, or Process and select desired metrics to view their graphs.
On a graph, the time granularity varies with the monitoring period and aggregation type. For details, see Aggregation Types and Time Granularities for Different Monitoring Periods.
For details about the aggregation methods supported by Cloud Eye, see What Aggregation Methods Does Cloud Eye Support?

You can enable auto refresh and set the refresh interval to 10 seconds, 1 minute, 5 minutes, or 20 minutes.
Hover your mouse over a graph. In the upper right corner, click to enlarge the graph for detailed data.
In the upper right corner, select a monitoring period (Last 15 min, Last 30 min, Last 1h, Last 2h, Last 3h, Last 12h, Last 1d, Last 7d, or Last 30d), or select Select Range to customize one, to view historical monitoring data from the last 155 days.
In the upper right corner of the graph, click to create an alarm rule for the metric. For parameter details, see Creating an Alarm Rule (New Version).